This desktop application addresses the challenge of organizing and enhancing long YouTube video transcripts by leveraging AI. It targets users who need to process extensive video content, such as researchers, students, or content creators, providing a structured and readable output format suitable for further analysis or consumption.

How It Works

The application fetches transcripts from YouTube playlists or individual videos. It then segments these transcripts into manageable chunks, processing each chunk with Google's Gemini API. A context prompt, including previously refined text, is passed to the API to maintain coherence. This iterative chunk-based refinement, configurable by the user, allows for detailed processing of lengthy content, ultimately producing a single, enhanced markdown file.

Quick Start & Requirements

• Install dependencies: pip install -r requirements.txt
• Run the application: python main.py
• Requires Python 3.9+ and a Google Gemini API key.

Highlighted Details

• Multiple AI refinement styles: Balanced, Summary, Educational (with glossary), Narrative Rewriting, and Q&A generation.
• Supports single video URLs and configurable chunk sizes for API calls.
• Outputs to formatted markdown files, compatible with tools like Obsidian and NotebookLM.
• User-friendly PyQt5 graphical interface.

Maintenance & Community

The project appears to be a personal project with a "Buy me a coffee" link for support. No specific community channels or contributor information are listed in the README.

Licensing & Compatibility

The README does not explicitly state a license. It is a Python application, and its compatibility with commercial or closed-source projects would depend on the underlying Gemini API terms of service and any unstated project license.

Limitations & Caveats

The effectiveness of the AI refinement is dependent on the quality of the YouTube transcripts and the Gemini API's performance. The "free" nature of the Gemini API may be subject to Google's usage policies and potential future changes.